http://www.defmacro.org/ Ramblings on Lisp, Haskell, C++, compilers, web frameworks and more. en-us http://www.defmacro.org/ramblings/relational.html Explicitly stating an interest for non-relational technologies is a little misleading. It suggests contempt for relational databases. And the talk about innovation suggests RDBMSes are are not innovative. Perhaps I started off on the wrong foot. Although my calculus is rusty, I am a mathematician at heart, and I can only admire the beauty and simplicity of relational algebra. Although modern databases are "pragmatized" versions of relational algebra, with much of its mathematical beauty stripped away, they appeal to me far more than most other tools. It's not just relational algebra. There are other things going for traditional databases in terms of mathematical beauty and innovation. For instance, it is far from obvious that it is possible to implement all four ACID properties simultaneously, and occasionally I am still amazed at how lucky we got. It is also amazing that traditional RDBMSes are interactive, while almost all other mainstream tools are not. I can type queries, compile stored procedures, modify the schema, and get immediate results, all without restarting the database. I am still fascinated at how people take this functionality for granted but don't mind complete impotence of Java in this department (sorry, hot deploy does not melt my butter the way Query Analyzer does). So what's missing? Why venture out to play with alternatives? Thu, 07 Aug 2008 02:13:12 EST http://www.defmacro.org/ramblings/relational.html http://www.defmacro.org/ramblings/fractal.html Fractal is a Direct3D demo I wrote a number of years ago to showcase real-time procedural generation of natural phenomena on modern GPU hardware (GeForce 2, at the time). The demo implements the four elements - earth, fire, air (clouds and wind), and water. Procedural generation of natural phenomena was my passion before Lambda Calculus took over my life. Wed, 02 Apr 2008 02:44:53 EST http://www.defmacro.org/ramblings/fractal.html http://www.defmacro.org/ramblings/s-query.html One downside of using Common Lisp to get things done is that occasionally libraries taken for granted in other languages are missing. I was using the excellent S-XML library to convert XML to s-expressions, when I realized that Common Lisp has no library equivalent in functionality to an implementation of XQuery. I needed to extract quite a bit of information from the s-expressions I acquired. Using CAR, CDR, MEMBER, and FIND quickly became infeasible. I had to roll my own query mechanism. It took me about twenty minutes to produce an acceptable solution. It's only around 45 lines of code. Tue, 01 Apr 2008 22:11:17 EST http://www.defmacro.org/ramblings/s-query.html http://www.defmacro.org/ramblings/lambda-reducer.html About a month ago I decided to buy some time from society by doing a trial run of graduate studies - I took some Computer Science courses at Stony Brook University. The first homework for the Programming Languages course was to write a lambda calculus reducer - a task I happily jumped on. A few hours later I had it implemented in 105 lines of Haskell code. Sun, 24 Feb 2008 16:55:04 EST http://www.defmacro.org/ramblings/lambda-reducer.html http://www.defmacro.org/ramblings/ui-dsl.html I like HTML in the same way I like PDF - as a document serialization format that does a reasonable job and that I never want to modify by hand. This is one of the main reasons I started Weblocks framework - I never wanted to write a line of HTML again. Ironically, I ended up writing a lot of HTML and learning more about its quirks, accessibility issues, and CSS hooks than I ever wanted to, but I finally ended up with a high level user interface definition language embedded into Common Lisp. Finally, HTML is out of my life, for good. Being lazy does pay off. Wed, 13 Feb 2008 21:32:43 EST http://www.defmacro.org/ramblings/ui-dsl.html